摘要
[目的/意义]由于汉语文本缺乏语义结构化,机器无法理解自然语言。因此,对汉语文本的语义结构化挖掘,有助于用户对其进行深层次利用和创新,从而推动数字人文研究发展。[方法/过程]文章以历史事件文本为研究对象,采用语义网技术,通过优化时间概念的语义表达方法,构建历史事件本体知识模型KMHEO;借助外部知识库,采用数据关联技术,将非结构化和结构化数据转换成遵循统一标准的结构化数据;并在此基础上,从单术语、双术语和多术语三种检索模式角度,设计了一个历史事件本体应用原型检索系统。[结果/结论]实验结果表明,文章设计的历史事件本体应用原型检索系统可以为用户提供形式多样的知识查询模式,能较好地解决历史事件中存在的关联缺失问题,为历史现象分析研究提供了新思路。
[Purpose/significance]Due to the lack of semantic structure in Chinese text,machines cannot understand natural language.Therefore,the semantic structure of Chinese text will help users to make deep use and innovation of it,which promoting the development of digital humanities research.[Method/process]This paper takes historical event text as the research object,adopting semantic web technology,and optimizing the semantic expression method of time concept to construct historical event ontology knowledge model(KMHEO).With the help of external knowledge base,adopting linked data technology to convert unstructured and structured data into structured data that conforms to unified standards.And on this basis,a historical event ontology application prototype retrieval system is designed from the three retrieval modes of single term,double term and multi-term.[Result/conclusion]The experimental results show that the prototype retrieval system of historical event ontology designed in this paper can provide users with a variety of knowledge query modes,which can better solve the problem of missing associations in historical events and provide new ideas for the analysis and research of historical phenomena.
出处
《情报理论与实践》
CSSCI
北大核心
2021年第6期171-179,共9页
Information Studies:Theory & Application
基金
国家自然科学基金面上项目“关联数据驱动下我国非遗文本的语义解析与人文计算研究”(项目编号:72074108)
南京大学文科青年跨学科团队专项“面向人文计算的方志文本的语义分析和知识图谱研究”(项目编号:2020300093)的成果
江苏青年社科英才和南京大学仲英青年学者等人才培养计划的支持。
关键词
关联数据
历史事件
本体知识模型
知识图谱
语义检索
linked data
historical event
ontology knowledge model
knowledge graph
semantic retrieval
